Abstract:The recursion hypothesis made by Hauser, Chomsky & Fitch(2002) claimed that the faculty of language in the narrow sense comprised only the computational mechanism of recursion and was uniquely human.The experiment on cotton top tamarins (Fitch & Hauser 2004) supported the hypothesis, whereas the experiment on European starlings (Gentner et al., 2006) disproved it.Nevertheless, the two experiments have methodological deficiencies in using meaningless artificial phonetic strings of the forms (AB) ²and A ²B ²as training stimuli, as there is uncertainty about segmenting the strings into discrete units, and corresponding grammatical structures.Therefore it is impossible to determine whether the animals have learned a context-free grammar(CFG) or not.The tests on native speakers of Chinese using Chinese words consisting of syllable strings of the forms (AB) ²and A ²B ²show that they mainly used double iterations strategy for processing the two patterns rather than a recursive, center-embedding, CFG.